Why Sample Matters
October 21, 2021
Your latest Nielsen survey is released, and the first thing you do is check your station’s performance. It doesn’t matter whether you’re in a PPM or a diary market; how your station fared is the most important element of the Nielsen survey.
However, many broadcasters do not take the time to examine how well Nielsen conducted their survey. When they do, it is usually only following a down book, and they’re looking for a reason to blame Nielsen. In other words – shoot the messenger.
Regardless of your station’s results in any survey, every broadcaster – whether they are in sales, programming, or management – should be aware of how well Nielsen performed.
To stay on top of things, we recommend that you assess Nielsen’s sample performance for every single survey. If you don’t understand how effectively a survey has been conducted, you could be making bad decisions based on either false positives or false negatives.
The first and probably most important piece of data to look at is sample size. Typically broadcasters look at both the overall sample (Persons 6+ or 12+) and the sample in your broad target demo (e.g. Adults 25-54). How does this sample compare to previous surveys and Nielsen’s sample target? Be aware that in diary markets, the sample represents how many total diaries from the entire survey period were used in the estimates. For PPM markets, Nielsen provides both the total number of in-tab meters from the entire month and the number of in-tab meters that factored into the ratings on an average day of the survey.
Generally speaking, the more sample the better. It is also important to look at how well the sample represents the population in your market. To determine this, look at the proportionality. A proportionality index of 100 means that the percent of the sample for that portion of the population matched the percent of the population.
Typically we look at Nielsen sample and proportionality by:
- Demographic cell
- Ethnic group
Proportionality below 100 results in the average meter or diary being weighted up. Keep in mind that lower proportionality in a particular segment does not mean lower estimates for your station. When certain survey participants are weighted up, their listening is magnified. If you “catch” a meter or diary, they can be helpful to your ratings. The opposite is also true. Simply put, lower proportionality indices may result in greater fluctuation in the data. The takeaway is to be knowledgeable about how representative the Nielsen sample is of the population in your market, and how it may affect your station’s estimates.
To be fair, getting people to participate in a survey of any kind is difficult. Obtaining enough sample and good proportionality is not an easy task. They have to balance age, ethnicity, gender, and geography. Nielsen works hard to do just that. However, before you make decisions about your station’s programming, be sure you’re working with data that properly represents your market.
When was the last time you looked at Nielsen’s sample performance? If it’s been a while, do it now. If you don’t have time, let the Ratings Experts at Research Director, Inc. help. Reach out to us here.
-Charlie Sislen, Partner